There is science behind managing customers in lines that some banks ignore. The other day, at a national bank, there were separate teller lines with some of the lines of sight blocked from the teller’s view, and there were ropes between the lines to prevent customers from moving from one line to another. Unfortunately, for this bank, they could not have possibly designed it any worse. In this article, we look at the well-worn science behind queuing, show why the art of queuing is more complicated than most bankers think, and give some tips on how to optimize your line structure.
The Basics of Optimization
There are two basic ways banks handle queuing. One is to create multiple lines each going to a banker. The other is a single line that queues to multiple bankers. Which one is better?
Most banks know the science and choose to have a single line feeding multiple bankers. The classic study comes from mathematics and statistics professor Douglas Norton at Villanova University that found in a study of three individual tellers with three lines or a single line leading to three tellers; the single line was three times faster. The reason, of course, is that when you have separate lines the delay in one line impacts that whole line whereas the single line self-adjusts.
However, that is not the whole story.
Know What You Want To Optimize
Most traditional studies were conducted with the goal of finding the queue structure that optimizes to the shortest wait times. However, there is customer satisfaction to also take into account. It turns out that a longer, single line creates less satisfaction to start. Seeing a long line lowers satisfactions as some customers see a line and choose to use the ATM or to come back. In separate studies that we have conducted, this occurs whenever there are three or more people in a line, no matter what the structure.
If your goal is to optimize wait times and customer satisfaction and your customers consistently leave because they see a long single line, then multiple lines might be better.
However, given a choice between continuously moving down a nine-foot line or inching three feet in three separate lines, those moving tend to have higher satisfaction. Thus, if the goal is to reduce wait times and optimize satisfaction for those customers that choose to wait in line, a single line is better.
When To Have Separate Lines
It turns out that humans are amazingly sensitive and good at equalizing lines. Many academic studies that show that a single line is faster had the unrealistic constraint that participants were in line at one time. This is rarely the reality, and the more realistic scenario is that customers come in over a period of time and get in line. If this is the case, and customers have the ability to move from line to line in case one is slower, then the difference in wait times between the two structures is negligible.
Supermarkets, for instance, have this setup and use multiple lines as the lines are easily self-adjusting. Also, long single lines often block traffic paths; multiple, shorter lines are sometimes the only option as blocking traffic patterns end up decreasing satisfaction for everyone.
Most all academic studies on wait times and satisfaction have ignored the employee. Since employees are banks’ most valuable assets, it is probably worth taking optimizing along this dimension as well. Here, it turns out that employees generally have more satisfaction when they are responsible for their line providing they can see their line.
In a study by the University of Washington and Purdue University, creating visual obstructions hurt both wait times and satisfaction for all parties. They also found that when there are no visual obstructions, lines were self-equalizing, customers were free to move between lines, AND transaction was homogeneous as to time, then parallel lines were faster than a single one. The visualization of the lines leads to immediate feedback, and at some level, competitiveness which drives up the feeling of productivity for the employee while shortening wait times and presumably increasing satisfaction for the customer.
There is one other aspect that is rarely discussed in banking – the outcomes of transactions. If your wire time cut-off is at 3 pm and long wait times cause customers to miss the cutoff, then all metrics – wait times, customer satisfaction, employee satisfaction, and outcomes suffer.
This problem is particularly acute in healthcare settings when both single or multiple lines would be sub-optimal as not all customer cases are the same. It doesn’t matter how short the wait time or satisfaction level is for that customer with a nosebleed if the customer behind them dies from poisoning while waiting.
A 2013 joint study from Harvard and Brandeis University in combination with Kaiser Permanente found that a managed, dedicated queuing system optimized along most dimensions. Being able to move customers around to separate lines based on their need and urgency while having a main, single-line, pooled queue ends up being the best structure.
If all that sounds a little too complicated for a bank, it is important to note that it is this very system that Bank of America and other banks often use in their main metro branches. Having a concierge to help manage customers to the main teller line, to product-specific desks or to universal bankers have shown a high level of overall success.
Putting This Into Action
Before you keep things status quo in your offices and branches, it pays to give some thought to your physical layout, the types of transactions being conducted, traffic flow and your banker capabilities to decide on the best-queuing structure. Many banks haven’t thought about this in 30 years despite huge changes in traffic and types of transactions.
All this said, if none of this is a problem for you because you don’t have long enough lines in your branches, then it might be time to expand your mobile banking capabilities.
Submitted by Chris Nichols on May 07, 2018